Hypotheses and projections are constantly refined. But in comparison to estimates on things like coal or climate, the population growth models are almost trivially simple, and while the numbers are adjusted up and down regularly to account for actual data, all data we have show the population growth consistently slowing down.
For starters, one of the things that make the population models straightforward compared to a lot of other things we might try to model, is that we can compare countries, and as it happens there are patterns that have consistently applied to countries as they develop, and that we have detailed data on: As life expectancy increases, growth rates drop to near or below maintenance rates.
Unless the remaining countries with rapid growth are somehow drastically different, it would take massive, earth-shattering changes for the population growth not to stop. It's not realistically a question of if it stops, but when and how high the peak will be, and how much the population will fall back afterwards before growth resumes (the expectation is that it will fall back because we get a "bulge" similar to what we're seeing these days due to the baby-boomer generation, and eventually the people in that bulge will start to die off).
Even many developing countries have long ago entered into the phase where it is not birth rates that is the cause of ongoing growth, but improving life expectancy, which means that their growth will eventually stop.
And with countries like China heading rapidly towards contraction as early as 2030 (with net growth now down to around 1/3 of its peak in the mid 80's, driven by a fertility rate far below maintenance), even the remaining high growth countries would have to dramaticall raise their population growth if we are to continue seeing overall growth.
- The growth rate is slowing. The growth in absolute terms is still sufficient to make the line look near linear when you look at it over such a long time span, especially given that life expectancy is also increasing almost everywhere (see my last point below).
- The growth rate is fairly low. E.g. around 2005, the growth rate was about 1.1%, and the decline is slow (e.g. it's projected to get to around 0.33% or so around middle of the century), so seeing the change on a graph that plots billions of people against a period of decades gets tricky.
- There are large temporal distortions due to changes in life expectancy, and the size of generations. To give an extreme example: Consider if a population to begin with was static - births and deaths were perfectly matched. Now consider if this population stopped having children, yet at the same time, everyone started living 40 years longer. In this (totally impossible of course) scenario, it would take 40 years from the birthrate collapsed until the population size would start dropping. While something that extremely obviously would not happen, less extreme variations are. E.g. some areas of China has a birth rate of well below 1 child per woman, and the country as a whole is well below 2, while you need somewat more than 2 to maintain a population (to account for men + people who never procreate), but the population is still increasing because population is still young on average due to the massive population growth in the 50's and 60's coupled with rapidly increasing life expectancy.
This last point means that we're still seeing a combination of the effect of birth rates going as far back as at least 1950's and all increases in life expectancy since, that correctly reflects current growth, but gives us a very distorted idea of where the population size is headed.
I looked at your graph, and I was wondering about that too. I'm no expert, but consider this.
Population growth is usually expected to be exponential. The fact that the graph looks linear we could interpret as a positive sign.
Also, if we check out the Earth's growth rate [1], the percentages have indeed slowed down. The left part (1965 - 1970) peaking at 2.10% looks like Generation X, the children of the baby boomers. The latest datum at 2012 returns 1.15%, and is still trending slightly negative.
The growth rate charts become more interesting as you isolate specific regions and countries. Sub-Saharan is the only region which is mostly trending positive [2]. Japan has received some press in recent years because its growth rate has actually crossed the x-axis[3] .
Hypotheses and projections are constantly refined. But in comparison to estimates on things like coal or climate, the population growth models are almost trivially simple, and while the numbers are adjusted up and down regularly to account for actual data, all data we have show the population growth consistently slowing down.
For starters, one of the things that make the population models straightforward compared to a lot of other things we might try to model, is that we can compare countries, and as it happens there are patterns that have consistently applied to countries as they develop, and that we have detailed data on: As life expectancy increases, growth rates drop to near or below maintenance rates.
Unless the remaining countries with rapid growth are somehow drastically different, it would take massive, earth-shattering changes for the population growth not to stop. It's not realistically a question of if it stops, but when and how high the peak will be, and how much the population will fall back afterwards before growth resumes (the expectation is that it will fall back because we get a "bulge" similar to what we're seeing these days due to the baby-boomer generation, and eventually the people in that bulge will start to die off).
Even many developing countries have long ago entered into the phase where it is not birth rates that is the cause of ongoing growth, but improving life expectancy, which means that their growth will eventually stop.
And with countries like China heading rapidly towards contraction as early as 2030 (with net growth now down to around 1/3 of its peak in the mid 80's, driven by a fertility rate far below maintenance), even the remaining high growth countries would have to dramaticall raise their population growth if we are to continue seeing overall growth.